Did These Researchers Just Solve One of the Mysteries of Getting High?

The cannabis world can be an uncertain place. Often, a lack of understanding comes from a dearth of clinical research into cannabis studies. But many countries around the globe have increased cannabis analysis in recent years — and now, there’s hope that the U.S. will soon enough increase its research frequency, that is if the plant is rescheduled to a Schedule III substance, per the Department of Health and Human Services’ August 2023 recommendation to the DEA.

However, as it currently stands, research continues to lag. Two critical areas impacted by the analysis gap are dosing and public safety. For the former, patients and medical professionals alike often struggle to find ideal cannabis dosages, leaving most to play guess and check. For the latter, concerns over impaired driving are common, and have held up legalization in several states. Without a breathalyzer-type device to gauge a person’s THC levels, it’s impossible to know if or how high a person is when stopped behind the wheel — making lives difficult for drivers and law enforcement. 

In early September, researchers Vikram Narayanan Dhamu, Nathan Kodjo Mintah Churcher, and Dr. Shalini Prasad of the University of Texas at Dallas announced the creation of CannibiSenS, a handheld saliva-based rapid test capable of accurately measuring a person’s THC levels 94 percent of the time. The device took inspiration from at-home glucose readers used by diabetics. This technology makes significant strides in cannabinoid detection with its capabilities to separate readings for THC, a psychoactive cannabinoid, and CBD, a nonpsychoactive cannabinoid. 

Its innovation has led to speculation that both dosing and public safety concerns could soon lessen. In an interview with Rolling Stone, Dr. Prasad said that while the innovation may one day play a part in impairment tests, her team sees its most promising potential in the medical space, helping people find their unique dosage. But before the device can be put to use in any space, Dr. Prasad said it needs 12 to 18 months more research to better understand its capabilities — and strategic partners to further aid in its development.  

Can you explain how CannibiSenS came about?
It actually started in 2018 when we were trying to understand if there were analytical tools out there that help us understand the metabolism of cannabis. It all came about because I was working, at that time, studying alcohol and alcohol metabolism. 

Now, alcohol is very volatile and gets out of your body quickly. Law enforcement just looks at the blood alcohol content, but actually, if you have repetitive alcohol consumption, you create a metabolite, which is called ethyl glucuronide (EtG), and that’s how alcohol is metabolized. Based on the concentration of EtG expression, you end up assessing, frequency of use or consistent, sustained use of alcohol. 

So from that, I kept thinking, ‘Is there an opportunity now, especially if there’ll be sustained consumption, for people to understand what that means? Everyone’s metabolism is different. It matters to have an objective read. 

We found that using electroanalytical controls, we can get a direct measure, especially of the psychoactive component, Delta-9-THC. The problem was the other non psychoactive component (CBD). Over time, especially during the pandemic, a lot of work went into saliva testing. So we learned from that and readapted it to our platform. 

So it really was just about trying to simplify cannabinoid readings and making it accessible for everyone? 
Yeah, the way I see this is, we all try to hack our own bodies, right–understand how we are metabolizing things and what’s happening. That’s what a continuous glucose monitor does. It exactly tells you how your glucose is. 

You don’t need to be a diabetic to want to understand your metabolism and improve peak performance, right? The amount of cannabis you’re consuming matters, especially in the medical case. Those are the kind of things that interest me in trying to understand, solve and build objective tools that help people better understand what’s happening.

When it comes to dosing, most industry advice boils down to start with a low dosage, wait 30-60 minutes and see how you feel. Is CannibiSenS intended to improve upon this method? 
I see it as the first step. That’s why it needs to be a self-monitoring device–something that they could use at home. They don’t need to collect and ship off the sample somewhere and get a lab analytical grade output. 

That’s how I have envisioned the utility of this platform. Because truly understanding metabolism, I think, is the key .

How do other minor cannabinoids and terpenes factor in? Is this a factor in CannibiSenS?  
This is an ongoing effort. [Cannabis consumption] is actually a very complex and heterogeneous response profile. If you can add them all in, then we can truly give you a total profile of psychoactivity and when the signal will start to drop.

We’re not yet there, but you are kind of, in a sense, reading my mind.

With a 94 percent accuracy, is that enough to qualify as a viable DWI test option, or are improvements required?
More than accuracy, we have to focus on precision. Can we hit that 94 percent consistently across the heterogeneity in a population? That has to be resolved. And that level of precision in the testing is important as long as you can maintain that accuracy of the 94 percent.

Yes, there is that error of six percent. How do you factor that out? Is that something that would require a second test? Where exactly is that error coming from? Resolving those issues are critical if such a test has to be adopted for something like a DWI. 

You have to build precision into it. But this is the first device that actually shows promise. Because in reality, you’ve done nothing to dilute the saliva. You’re using it exactly as it is. And that means that realistically, that sensor interface has both the sensitivity as well as the specificity for the target. 

Now it’s a matter of engineering in intelligence to be further able to resolve those thresholds. So, I see this as an opportunity, but I don’t see that as the only use case. Ultimately the tool is agnostic to the application and you want this to work for whatever use.I think most of the need right now first is in medical use. 


I think some people will see this news and immediately say, ‘They found a cannabis DWI test and everything is solved now.’ 
It’s not something I decide. It’s not a mouse trap at all. That’s not the purpose. And perhaps if there is a need and there is all the regulation that says you have to build that, then this could be used in that space.

What’s been the response since the news was announced?
We’ve gotten support from the perspective of people wanting to build out the capability, scale it up. Those kinds of business folks are always interested. There’s of course a little bit of chatter, asking if the device is a mouse trap. Then there is a lot more, especially from researchers who are working with cannabis in the context of medical diseases, where you’re trying to understand what dose to give, especially to pediatric patients or to very vulnerable populations. Can we adapt and use this tool to find those questions and answers?